nharada

4 ms·

31.
by nharada 7mo ago
Surely this task must now be in the training data
32.
by nharada 7mo ago
Yeah as long as the chatbot is empowered to fix a bunch of basic problems I'm okay with them as the first line of support. The way support is setup nowadays humans are basically forced to be robots anyway, given a set of canned respons
33.
by nharada 8mo ago
Nice, I like the idea. It sounds like qualitatively you haven't had any performance regressions while doing this, but have you tested it at all on any sort of benchmark or similar eval? I'm curious how well the actual system perfo
34.
by nharada 8mo ago
Another interesting thing here is that the gap between "burned out but just producing subpar work" and "so crispy I literally cannot work" is even wider with AI. The bar for just firing off prompts is low, but the mental
35.
by nharada 8mo ago
+1 First, I agree with most commentators that they should just offer 3 modes of visibility: "default", "high", "verbose" or whatever But I'm with you that this mode of working where you watch the agent wor
36.
by nharada 8mo ago
This is awesome and I'm really happy to see this progress. Landing a new chemistry in a production car THIS YEAR is some crazy velocity, especially compared to where other Na-Ion batteries are in the development cycle elsewhere. Is any
37.
by nharada 8mo ago
That's a massive jump, I'm curious if there's a materially different feeling in how it works or if we're starting to reach the point of benchmark saturation. If the benchmark is good then 10 points should be a big improv
38.
by nharada 8mo ago
> worry that the US will fall behind the curve Man it's already over. It's hard to imagine the US autos EVER catching up at this point, even with state support.
39.
by nharada 9mo ago
I'm obviously biased, and I probably have more gripes than most about Waymo as a corporate entity, but the premise this article seems to be based on is "Waymo is a zombie company who will never release a real product" or some
40.
by nharada 9mo ago
I think the assumption is valid. Most of the reasoning components of the next gen (and some current gen) robotics will use VLMs to some extent. Deciding if a temporary construction sign is valid seems to fall under this use case.
41.
by nharada 10mo ago
In that case working at a startup would be a thing someone would only do as a last resort, and the talent pool would consequently be extremely low quality. Sounds damaging to the scene to me.
42.
Emergence of Human to Robot Transfer in VLAs(pi.website)
1 points by nharada 10mo ago|0 comments
43.
by nharada 10mo ago
It definitely feels like a jump in capability. I've found that the long term quality of the codebase doesn't take nosedive nearly as quickly as earlier agentic models. If anything it's about steady or maybe even increasing if
44.
by nharada 10mo ago
The general playbook in the US seems to be "do something that's bad for society but makes money, and eventually the people in charge will get enough FOMO that it'll be officially sanctioned as long as they get a cut".
45.
by nharada 10mo ago
I think we're at the point where both phone and SMS are such insecure and easily spoofed channels that we should basically not be using them for anything related to business or money. Maybe even for communication, given how easily scam
46.
by nharada 11mo ago
I don't really take your point here? You're saying you feel the author should justify the reasoning behind their moral position?
47.
by nharada 11mo ago
The end goal is to marry the lessons learned about HOW to learn in a virtual world with a high fidelity world model that's currently out of reach for this generation of AI. In a year or two once we have a world model that's realis
48.
Waymo robotaxis are now giving rides on freeways in LA, SF and Phoenix(techcrunch.com)
341 points by nharada 11mo ago|437 comments
49.
by nharada 11mo ago
I think the way China approached this is probably the better way -- heavily support companies in the commercial sector that can quickly iterate, invest heavily to improve the tech, and scale up manufacturing. They'll always have the la
50.
by nharada 1y ago
No matter how fast you run, you won't win a race where you're just going back and forth nonstop. It really does feel like the US is completely hosed when it comes to energy (and thus, industrial relevance broadly). Every 4 years w
51.
by nharada 1y ago
You're right, I can't find any primary sources for this number. Yahoo[1] reports this number and attributes it to Bloomberg NEF but I can't find an actual article from Bloomberg with this number, or any actual target number i
52.
by nharada 1y ago
You're right, I can't find any primary sources for this number. Yahoo[1] reports this number and attributes it to Bloomberg NEF but I can't find an actual article from Bloomberg with this number, or any actual target number i
53.
by nharada 1y ago
Oh yeah I get what you’re saying but is that a thing? Like CATL has had lines before that aren’t making a profit?
54.
by nharada 1y ago
What makes you say that? I don’t know this space very well
55.
by nharada 1y ago
CATL is claiming mass production of their sodium-ion batteries starts in December, with a target price of $10/kWh. If that ends up even partway true it'll completely change the economics of power storage.
56.
by nharada 1y ago
> Price is the primary decider That's not even true for normal bicycles. Serious cyclists can spend massive amounts of cash on their rides. Aside from that, if you're using this as a vehicle you want more than just the cheapest
57.
by nharada 1y ago
https://en.wikipedia.org/wiki/Skyscraper_Index
58.
by nharada 1y ago
How come I never see any concrete proposals for how to equitably distribute the wealth of AI? It's always either "stop AI immediately for the sake of our labor" or "don't worry sometime in the future everyone will l
59.
by nharada 1y ago
http://archive.today/HHhEy
60.
America's Tech Right Is Obsessed with Building Giant Statues(bloomberg.com)
6 points by nharada 1y ago|1 comments

More ›